Data Science Commons^™

Utilizing Physics Informed Neural Networks For Disrupted Signal Dynamics, Nicholas J. Joyner

Electronic Theses and Dissertations

Physics-informed neural networks (PINNs) have been used in many applications including engineering and physical sciences. PINNs allow the incorporation of a priori understanding of a process’ structure into the modeling. We attempt to leverage the PINN structure toward the evaluation of disruptions to classical dynamical models by combining elements of ordinary differential equations into our loss function with sigmoidal gating to balance the penalties for deviations from the data with those for structural deviations. This enables the identification of the signal structure and the limits of disruption influence. As a use case, we consider stock value from 2019-2021, which expresses …

Addressing The Problems Of Data Variations, Quality, And Scarcity In Training Deep Neural Networks, Jian Sun

Electronic Theses and Dissertations

The performance of deep neural networks (DNNs) is strongly influenced by the characteristics and quality of the underlying datasets. This Ph.D. dissertation addresses three pervasive data challenges-imbalance, quality degradation, and scarcity-that commonly hinder the effectiveness of DNNs in computer vision (CV) and natural language processing (NLP) applications.

Class imbalance remains one of the most frequent causes of degraded model generalization. While Focal Loss effectively mitigates inter-class imbalance by assigning higher weights to minority classes, it struggles with intra-class imbalance, particularly in video datasets where longer clips dominate feature representation. To address this, I implement and utilize …

Topic Modeling And Culturomic Analysis Of 30,000 Books Over 100 Years Using Gensim, Michael A. Freeman

Electronic Theses and Dissertations

This thesis explores the cultural influence of historical events on English-language fiction published between 1820 and 1929. Using a corpus of 30,256 digitized books from Project Gutenberg, Latent Dirichlet Allocation (LDA) topic modeling was applied to identify recurring themes across eleven decades. The study sought to determine whether historically significant events could be detected within fictional narratives. One clear instance emerged: Napoleon Bonaparte and the Napoleonic Wars appeared explicitly in the 1820s corpus. Beyond this, several thematic patterns were observed—such as maritime language in the 1840s, national identity in the 1880s, and youth-oriented dialogue in the early 20th century—that plausibly …

Stochastic Functional Data-Driven Models For Real-Time Battery Health Forecasting Under Dynamic Operating Conditions, Joshua Owusu

Electronic Theses and Dissertations

This thesis provides an effective statistical model to predict the real-time state of lithium-ion batteries for reliable Battery Management Systems (BMS). It highlights battery data (voltage, current, temperature) as smooth functional curves. The principal method demonstrates diminishing trends to health outcomes like State of Health (SoH) and Remaining Useful Life (RUL) by employing Functional Principal Component Analysis (FPCA) and Bayesian Functional Linear Models (FLMs). The primary objective is to figure out how uncertain forecasts are. Simulations demonstrate that the highest accuracy (lowest MSE) is achieved through low noise levels along with large sample sizes. The final system provides a highly …

Postprocessing Gan-Generated Synthetic Time Series Using Dynamic Time Warping, Md Raisul Islam Roni

Electronic Theses and Dissertations

Generative Adversarial Networks (GANs) are a class of deep learning models capable of producing realistic synthetic data that preserve the statistical and temporal characteristics of real datasets. The DoppelGANger (DGAN) framework extends this approach to time series data by jointly modeling temporal dependencies and contextual metadata. However, synthetic sequences generated by GAN may show temporal misalignment, resulting in inconsistencies when compared with real data. This study presents a postprocessing framework based on Dynamic Time Warping (DTW) and its differentiable extension Soft-DTW to improve the temporal alignment of synthetic time series. The framework is evaluated using quantitative measures of alignment and …

A Leslie System For A Demographic Simulation: From An Actuarial Point Of View, David Kings

Electronic Theses and Dissertations

This thesis develops a discrete stochastic linear systems interpretation of age–stage demographic evolution grounded in Leslie operators and realized in a discrete-event simulation implemented with salabim. The central claim is that one annual cycle of the simulation constitutes a cone-preserving, stochastic affine transformation on a high- dimensional population state vector indexed by age, sex, marital status, household type, employment, and education, and that the composition of yearly operators yields a random matrix product whose top Lyapunov exponent is the stochastic counterpart of the Perron–Frobenius growth rate (Caswell, 2001; Tuljapurkar, 1997)[1, 2]. The actuarial bridge is constructed by mapping simulated survival …

A Persistent Homology Framework For Scrna-Seq: Assessing Clustering Robustness And Quantifying Preprocessing And Integration Effects On Topological Features., Jonah Daneshmand

Electronic Theses and Dissertations

As single-cell RNA sequencing (scRNA-seq) data expands, robust methods for integrating diverse datasets are critical. This dissertation applies Persistent Homology (PH), a technique from Topological Data Analysis (TDA), to a collection of scRNA-seq datasets spanning eight tissue types to quantify how data integration affects topological features and biological interpretability. We assessed global topological structure using Betti curves, Euler characteristics, and persistence landscapes across raw, normalized, and integrated data representations. Our analysis revealed a performance inversion: while conventional methods excelled on unintegrated data, high-granularity topological methods, particularly those sensitive to global data structure, became superior after integration. This suggests a synergy …

Grokking Applied To Chaotic Iterates Of The Logistic Map, Felix Donkoh

Electronic Theses and Dissertations

This thesis investigates grokking, the delayed transition from memorization to generalization in neural networks trained on deterministic chaotic data. Using an integer–arithmetic discretization of the logistic map, yn+1 =( a yn(p − yn))/ p 2 , bounded aperiodic sequences were generated across control parameters α ranging from 3.0 to 4.0. Transformer-based models displayed characteristic grokking curves. In periodic and chaotic regimes, validation accuracy rose suddenly after long plateaus, while at the Feigenbaum boundary (α ≈ 3.57) generalization failed completely. Increasing data diversity restored learning in chaotic domains, and explicit α–conditioning enabled a single network to generalize across all regimes. A …

An Income Subsystem As A Discrete Stochastic Leslie System: A Simulation-Based Approach, Fahd Nii Okantah Cobblah

Electronic Theses and Dissertations

This thesis formulates the household-income engine of an integrated population sim- ulator as a Discrete Stochastic Leslie System (DSLS). The nonnegative state vector nt ∈ Rk + aggregates income, savings, debt, employment, and transfers. (Here, the subscript + denotes the positive cone, i.e., vectors with nonnegative components). Annual evolution is linear in state, stochastic in coefficients: nt+1 = Ttnt + εt, with Tt : Rk + → Rk + cone-preserving. Exogenous macro drivers (inflation, employment, tax, salary inflation, mortgage) are forecast via ARIMA; forecasts multiply entries of Tt, preserving linearity in expectation while introducing realistic temporal correlation. The discrete-event implemented …

Estimation Methods For Bayesian Exponential Random Graph Models Under The Horseshoe Prior., Pamela Linares

Electronic Theses and Dissertations

Networks are powerful tools for modeling the complexity of social interactions, biological systems, and information spread. A leading statistical frameworks for analyzing network data are Exponential Random Graph Models (ERGMs), which provide a principled approach to capturing structural dependencies. However, ERGMs remain challenging to estimate, especially in sparse or high-dimensional settings where models suffer from degeneracy and unstable parameter inference. This paper proposes a penalized Bayesian approach to ERGMs that utilizes the horseshoe prior, a sparsity-inducing global-local shrinkage prior. This prior offers robust regularization while preserving important signals, improving estimation by shrinking irrelevant parameters and reducing the impact of extreme …

Evaluating Predictive Models For Predicting Total Score Of Beef Carcasses, Emmanuel Forson

Electronic Theses and Dissertations

The beef industry plays a vital role in global agriculture, with carcass quality and consumer preference being key determinants of market success. This thesis examines predictive modeling techniques for estimating the Total Score of beef carcasses, a composite measure representing yield and quality, primarily used by the Nebraska Cattlemen Association. Using data from the Nebraska Cattlemen’s Foundation Retail Value Steer Challenge (2000–2023), the study compares the performance of First Order Multiple Linear Regression (MLR) with three machine learning techniques: K-Nearest Neighbors (KNN), Random Forest, and Gradient Boosting Machine (GBM).

The analysis focuses on six key predictors: Hot Carcass Weight, Back …

Car Price Prediction Using Machine Learning: Analyzing The Dvm-Car Dataset, Yaman Abu Ghareebaih

Electronic Theses and Dissertations

The objective of this study is to predict car prices using machine learning models and the DVM-CAR dataset, which includes over 1.4 million images and car specifi- cations from 899 car models. Key factors such as mileage, engine power, and year of registration were analyzed for their correlation with car prices. Extensive data cleaning was performed, including filling missing values, identifying outliers, and normalizing numerical variables. Discrete variables like car make and body type were encoded using one-hot encoding. Linear relationships were analyzed with Multiple Logistic Regression, and Random Forest models were used for nonlinear patterns. Model performance was evaluated …

Investigating The Privacy-Utility Trade-O↵ In Synthetic Data Generation Using Correlated Attribute Mode, Kofi Sarfo

Electronic Theses and Dissertations

This thesis explores the privacy-utility trade-off in synthetic data generation using the Correlated Attribute Mode of DataSynthesizer, which employs Bayesian networks to model attribute dependencies. It focuses on integrating differential privacy mechanisms, particularly the Laplace mechanism, to inject controlled noise into synthetic data and enhance privacy protection. As organizations face challenges balancing data-driven decision-making with privacy regulations such as the General Data Protection Regulation and the California Consumer Privacy Act, synthetic data offers a solution by creating artificial datasets that preserve statistical properties while balancing data privacy and utility. This research investigates how different differential privacy parameters epsilon affect data …

Advancing Continuous Manufacturing: The Role Of Process Analytical Technology In Process Development, Samuel R. Henson

Electronic Theses and Dissertations

The pharmaceutical industry is actively pursuing technologies which improve manufacturing processes with the goal of producing high-quality pharmaceutical products for patients, manifesting in an industry-wide investment in continuous manufacturing (CM). Process analytical technology (PAT) has been recognized for its successful monitoring of critical quality attributes during routine production and is often cited alongside CM due to its ability to make timely, in-line measurements of intermediate materials. Various PAT tools are valuable in process development, particularly as continuous wet granulation processes are developed for use within pharmaceutical manufacturing. This work applied PAT and chemometric modeling during CM process development to enhance …

Exploring The Diagnostic Potential Of Radiomics-Based Pet Image Analysis For T-Stage Tumor Diagnosis, Victor Aderanti

Electronic Theses and Dissertations

Cancer is a leading cause of death globally, and early detection is crucial for better

outcomes. This research aims to improve Region Of Interest (ROI) segmentation

and feature extraction in medical image analysis using Radiomics techniques

with 3D Slicer, Pyradiomics, and Python. Dimension reduction methods, including

PCA, K-means, t-SNE, ISOMAP, and Hierarchical Clustering, were applied to highdimensional features to enhance interpretability and efficiency. The study assessed the ability of the reduced feature set to predict T-staging, an essential component of the TNM system for cancer diagnosis. Multinomial logistic regression models were developed and evaluated using MSE, AIC, BIC, and Deviance …

Interpretation Models For Prostate Lesion: Detecting, Explaining, And Understanding., Mehmet Akif Gulum

Electronic Theses and Dissertations

Prostate cancer is a major public health concern, affecting millions of men worldwide. While early detection and treatment of prostate cancer is critical for improving patient outcomes, the detection of prostate lesions is even more important for timely intervention and management of the disease. Prostate lesions are abnormal growths or lumps within the prostate gland, which may or may not be cancerous. The timely detection and accurate diagnosis of prostate lesions is crucial for effective treatment and management of the disease. In recent years, deep learning models have shown promise in accurately detecting and characterizing prostate lesions using advanced imaging …

Advancement Of Iterative Optimization Technology Algorithms Toward Calibration-Free Process Analytical Technology Applications, Adam Rish

Electronic Theses and Dissertations

The expansion of spectroscopic process analytical technology (PAT) tools within the pharmaceutical industry has the potential to elevate the current state-of-the-art of pharmaceutical manufacturing by offering opportunities for reduced quality testing times, enhanced process control, and greater production flexibility. Spectroscopic PAT tools are dependent on multivariate models to extract the relevant information from the spectral outputs. However, there is a substantial calibration burden for developing and maintaining these multivariate models that discourages the application of PAT, despite the encouragement from regulators. This has led to an interest in calibration-free methods such as iterative optimization technology (IOT) for spectroscopic PAT that …

Interpreting Shift Encoders As State Space Models For Stationary Time Series, Patrick Donkoh

Electronic Theses and Dissertations

Time series analysis is a statistical technique used to analyze sequential data points collected or recorded over time. While traditional models such as autoregressive models and moving average models have performed sufficiently for time series analysis, the advent of artificial neural networks has provided models that have suggested improved performance. In this research, we provide a custom neural network; a shift encoder that can capture the intricate temporal patterns of time series data. We then compare the sparse matrix of the shift encoder to the parameters of the autoregressive model and observe the similarities. We further explore how we can …

Multimodal Stylometry: A Novel Approach For Authorship Identification., Glory O. Adebayo

Electronic Theses and Dissertations

This dissertation introduces multimodal stylometry, a novel approach to authorship identification that integrates text and source code features for a comprehensive understanding of an author's unique style. Traditional stylometric methods have primarily focused on either text stylometry or source code stylometry, thereby neglecting the potential insights that multimodality may provide. This research aims to bridge this gap by proposing a framework that combines textual and source code data to enhance the accuracy and reliability of authorship identification. The study begins by reviewing existing literature on authorship identification and stylometry, highlighting the limitations of unimodal approaches. Leveraging recent advancements in multimodal …

Establishing “The Fossil Record”: A Database Of Vertebrate Paleontological Sites Across The State Of Tennessee, Sarah Mclaurine

Electronic Theses and Dissertations

Fossil localities across the state of Tennessee and the data related to those sites were compiled from Tennessee Division of Geology Bulletin 84, titled “Tennessee’s Prehistoric Vertebrates,” and stored in a Microsoft Access geodatabase housed by the Department of Collections at the East Tennessee State University Museum of Natural History located at the Gray Fossil Site. Included in the database are forms to enter new site localities, view information about those already entered, view and add data to a master faunal list for the state, view sites repository information and store and add documents that are key-word searchable from the …

Implementation Of Hierarchical And K-Means Clustering Techniques On The Trend And Seasonality Components Of Temperature Profile Data, Emmanuel Ogedegbe

Electronic Theses and Dissertations

In this study, time series decomposition techniques are used in conjunction with Kmeans clustering and Hierarchical clustering, two well-known clustering algorithms, to climate data. Their implementation and comparisons are then examined. The main objective is to identify similar climate trends and group geographical areas with similar environmental conditions. Climate data from specific places are collected and analyzed as part of the project. The time series is then split into trend, seasonality, and residual components. In order to categorize growing regions according to their climatic inclinations, the deconstructed time series are then submitted to K-means clustering and Hierarchical clustering with dynamic …

Convolution And Autoencoders Applied To Nonlinear Differential Equations, Noah Borquaye

Electronic Theses and Dissertations

Autoencoders, a type of artificial neural network, have gained recognition by researchers in various fields, especially machine learning due to their vast applications in data representations from inputs. Recently researchers have explored the possibility to extend the application of autoencoders to solve nonlinear differential equations. Algorithms and methods employed in an autoencoder framework include sparse identification of nonlinear dynamics (SINDy), dynamic mode decomposition (DMD), Koopman operator theory and singular value decomposition (SVD). These approaches use matrix multiplication to represent linear transformation. However, machine learning algorithms often use convolution to represent linear transformations. In our work, we modify these approaches to …

A Bridge Between Graph Neural Networks And Transformers: Positional Encodings As Node Embeddings, Bright Kwaku Manu

Electronic Theses and Dissertations

Graph Neural Networks and Transformers are very powerful frameworks for learning machine learning tasks. While they were evolved separately in diverse fields, current research has revealed some similarities and links between them. This work focuses on bridging the gap between GNNs and Transformers by offering a uniform framework that highlights their similarities and distinctions. We perform positional encodings and identify key properties that make the positional encodings node embeddings. We found that the properties of expressiveness, efficiency and interpretability were achieved in the process. We saw that it is possible to use positional encodings as node embeddings, which can be …

A Data-Driven Multi-Regime Approach For Predicting Real-Time Energy Consumption Of Industrial Machines., Abdulgani Kahraman

Electronic Theses and Dissertations

This thesis focuses on methods for improving energy consumption prediction performance in complex industrial machines. Working with real-world industrial machines brings several challenges, including data access, algorithmic bias, data privacy, and the interpretation of machine learning algorithms. To effectively manage energy consumption in the industrial sector, it is essential to develop a framework that enhances prediction performance, reduces energy costs, and mitigates air pollution in heavy industrial machine operations. This study aims to assist managers in making informed decisions and driving the transition towards green manufacturing. The energy consumption of industrial machinery is substantial, and the recent increase in CO2 …

Cannabidiol Tweet Miner: A Framework For Identifying Misinformation In Cbd Tweets., Jason Turner

Electronic Theses and Dissertations

As regulations surrounding cannabis continue to develop, the demand for cannabis-based products is on the rise. Despite not producing the psychoactive effects commonly associated with THC, products containing cannabidiol (CBD) have gained immense popularity in recent years as a potential treatment option for a range of conditions, particularly those associated with pain or sleep disorders. However, due to current federal policies, these products have yet to undergo comprehensive safety and efficacy testing. Fortunately, utilizing advanced natural language processing (NLP) techniques, data harvested from social networks have been employed to investigate various social trends within healthcare, such as disease tracking and …

Do Integrated Circuits Make For An Integrated Supply Chain? A Network Analysis Of Trade Flows, Noah Martens

Electronic Theses and Dissertations

Integrated circuits (colloquially referred to as chips) are an increasingly critical commodity experiencing continuous and substantial rises in demand. These increases in demand recently resulted in shortages. This paper seeks to understand the market for chips and construct a framework by which a network analysis of trade flows can evaluate concentration in the international market for a particular product category. Leveraging this framework and proposing a new model, I evaluate the level and nature of concentration in the chips sector, as well as two key inputs to the manufacturing process, silicon and chip fabricators. I find moderate-to-high levels of concentration …

A Programmatic Geographic Information Systems Analysis Of Plant Hardiness Zones, Andrew Bowen

Electronic Theses and Dissertations

The Plant Hardiness Zone Map consists of thirteen geographical zones that describe whether a plant can survive based on average annual minimal temperatures. As climate change progresses, minimum temperatures in all regions are expected to change. This work programmatically evaluates predicted future climate projection data and converts it to United States Department of Agriculture-defined hardiness zones. Through the next 80 years, hardiness zones are projected to move poleward; in effect, colder zones will lose area and warmer zones will gain area globally. Some implications include changes in crop growing degree days, which could alter crop productivity, migration and settlement of …

Predicting High-Cap Tech Stock Polarity: A Combined Approach Using Support Vector Machines And Bidirectional Encoders From Transformers, Ian L. Grisham

Electronic Theses and Dissertations

The abundance, accessibility, and scale of data have engendered an era where machine learning can quickly and accurately solve complex problems, identify complicated patterns, and uncover intricate trends. One research area where many have applied these techniques is the stock market. Yet, financial domains are influenced by many factors and are notoriously difficult to predict due to their volatile and multivariate behavior. However, the literature indicates that public sentiment data may exhibit significant predictive qualities and improve a model’s ability to predict intricate trends. In this study, momentum SVM classification accuracy was compared between datasets that did and did not …

Unsupervised Dimension Reduction Techniques For Lung Diagnosis Using Radiomics, Janet Kireta

Electronic Theses and Dissertations

Over the years, cancer has increasingly become a global health problem [12]. For successful treatment, early detection and diagnosis is critical. Radiomics is the use of CT, PET, MRI or Ultrasound imaging as input data, extracting features from image-based data, and then using machine learning for quantitative analysis and disease prediction [23, 14, 19, 1]. Feature reduction is critical as most quantitative features can have unnecessary redundant characteristics. The objective of this research is to use machine learning techniques in reducing the number of dimensions, thereby rendering the data manageable. Radiomics steps include Imaging, segmentation, feature extraction, and analysis. For …

The Role Of Generative Adversarial Networks In Bioimage Analysis And Computational Diagnostics., Ahmed Naglah

Electronic Theses and Dissertations

Computational technologies can contribute to the modeling and simulation of the biological environments and activities towards achieving better interpretations, analysis, and understanding. With the emergence of digital pathology, we can observe an increasing demand for more innovative, effective, and efficient computational models. Under the umbrella of artificial intelligence, deep learning mimics the brain’s way in learn complex relationships through data and experiences. In the field of bioimage analysis, models usually comprise discriminative approaches such as classification and segmentation tasks. In this thesis, we study how we can use generative AI models to improve bioimage analysis tasks using Generative Adversarial Networks …